Currently looking for HGSS and going mad. Ok, I understand cannot be a guarantee progression of them, but surely if you are in the right place and there say a 5% chance per minute, by 20 minutes it should be a 100% chance? So you can get lucky and beat the timer but if you aren't lucky at least you know there is a light at the end of the tunnel!
Why doesn't RNG scale?
- Thread starter Pendergast
- Start date
Two words : binomial probability
Because it's only a three letter acronym.
Here's the math. The chance of it the thing NOT spawning in one minute is 95%. The chance of it not spawning in 2 minutes is 95% X 95% (assuming they are independent, which I believe they are). After 20 minutes, you're still at about 36% chance of the thing not happening. The probability never actually gets to 100% it will happen (equally, 0% it wont), but for it to fall to below 1% chance of not happening, you'd need to wait 90 mins.
Good explanation, what I was trying to get at however is that they FORCE it to be 100% after a certain time. Have a timer for events in your area, and if you crap out before they expire at least when they do the event is guaranteed to happen.
I was looking for them to artificially force it however.
RNG is bad enough, but the idea that the game play is go somewhere and wait for 20 minutes, would, I suspect, lead much hilarity (not) on the forums. At least RNG give you some false hope.
Why cant we just have a deterministic way of getting stuff. An absolutely guaranteed "do this and get that" means to get all the different things we need in the game?
EDIT: This doesnt mean the "do this" part needs to be an easy task.
Why cant we just have a deterministic way of getting stuff. An absolutely guaranteed "do this and get that" means to get all the different things we need in the game?
EDIT: This doesnt mean the "do this" part needs to be an easy task.
Last edited:
verminstar
Banned
The fact its random means it cant be predictable...if it scaled, it wouldnt be random and unpredictable and wouldnt be called rng in the first place ^
![[Diplomacy] Carnegie](https://forums-cdn.zaonce.net/data/avatars/m/123/123809.jpg?1553527149){kind=avatar}
Nah, they threw a dart at a dictionary and picked the word it stuck in.
It wouldn't be 5% anymore then.
Last edited:
Ok, I see everyone focussing on random; can't it be deterministic within a overall basket of time and random within that basket? 20 minutes is the time basket, can pop up randomly within that time period but if it doesn't a switch is thrown and you get the result you were waiting for. They obviously don't want a fast drop rate otherwise would increase the overall chance so while this keeps the overall rate low, it removes the frustration of hours of hunting for something that should be right there. Anyway, not going to do it I imagine. Just very frustrating.
However you cut it, waiting is not very engaging gameplay.
To be fair, some games do have some kind of normalised RNG that enforces the average even at the expense of making the % chances adapt to the situation.
I know League of Legends used to do this for critical hits, where a 25% critical meant not that there was a 25% of each hit being a critical, but that 25% of your hits would be criticals. Every attack that wasn't a critical would increase critical chance, while every critical hit reduced said critical chance; there's still randomness and the potential range is the same, but the standard deviation is much lower.
I know League of Legends used to do this for critical hits, where a 25% critical meant not that there was a 25% of each hit being a critical, but that 25% of your hits would be criticals. Every attack that wasn't a critical would increase critical chance, while every critical hit reduced said critical chance; there's still randomness and the potential range is the same, but the standard deviation is much lower.
The question is, how does the game know you are specifically looking for HGSS? All rare items should have the same basic chance of appearing, but the probability of a specific rare is low and there's currently no way for the game to know what you are looking for.
One thing I've suggested (in the Suggest forum in the past) is that materials needed for a pinned blueprint are more likely to spawn. The only other real solution is to get rid of RNG, at least in a few places. That way, if you need HGSS, you'll know there is at least 1 place where you are guaranteed to find it.
Of course, the way FD thinks, that 1 place would probably be Hutton Orbital or Beagle Point!
I hope they will add some gameplay to finding USS's, like an active scan-mini-game with the discovery scanner. Then add something similar for planet's POIs and the DSS. There needs to be something more engaging, if not outright deterministic for finding those. Random tables are a quick, but unsatisfying filler for game content.
There's nothing wrong with using RNG.... IF that rng is backup by a failover mechanism..
If the rng is 1 in 5 and nothing else of seeing something, there is a change you will never see it. That is why, failover is required. Usually in any of the games I built, after x number of iteration, the random chance is reduced, to such time as on the next iteration, it is guaranteed, with no rng element.
FDev, don't use that, and rely on lazy coding to achieve their results. And, while it works some times, by pure definition of it being random, they lose control.
They just need to wake up and learn about game design to eliminate player frustration.
If the rng is 1 in 5 and nothing else of seeing something, there is a change you will never see it. That is why, failover is required. Usually in any of the games I built, after x number of iteration, the random chance is reduced, to such time as on the next iteration, it is guaranteed, with no rng element.
FDev, don't use that, and rely on lazy coding to achieve their results. And, while it works some times, by pure definition of it being random, they lose control.
They just need to wake up and learn about game design to eliminate player frustration.
Doesn’t the above maths depend on whether one check has a bearing on the preceding check?
I.e.
The likelihood of drawing the ace of spades from a deck of cards starts at 1 in 52 but the odds improve as each card is chosen and then discarded if it isn’t the ace of spades.
However
The odds of getting a heads when tossing a coin 52 times will always be 50:50. The previous coin toss has no bearing on the subsequent toss.
Without knowing how the RNG equation works (each check is independent of previous checks or is cognisant of previous results), you don’t know whether you are looking for a playing card or a coin toss.
Last edited:
I clearly stated my assumption, but yes, it is just that.
The reason I think they use random independent events is that they are inherently stateless. This makes the coding easier. If you go with the "looking for a card" approach, it means you need to store the state thing per player, and then have to deal with issues like when to reset things, what happens if multiple players join the same instance, eta.
The reason I think they use random independent events is that they are inherently stateless. This makes the coding easier. If you go with the "looking for a card" approach, it means you need to store the state thing per player, and then have to deal with issues like when to reset things, what happens if multiple players join the same instance, eta.
That's not how probabilities work.